This is on Netcup’s flash sale today and may be of interest to someone.
Processor: 8 vcores
Memory: 16 GB
Storage: 1,5 TB SSD-Performance
Traffic: 80 TB @ 1Gb/s port
Price: €14.49 (incl. 19% VAT)
Contract period: 1 month
Billing period: 1 month
3 Likes
Damn, those are some good specs. I would be interested to know what the CPU performance is like.
Their spring offer has 2 dedicated cpu cores and 20 Gb of Ram for € 9.99
Which one performs better?
Not easy to measure, as we don’t know how many clients share a vcore
Just picked one up. Here are the Geekbench results:
1 Like
Why DE can do these kind of crazy price?
Netcup Hetzner Contabo they all located at DE
Damn, thanks! What is the disk speed like?
Good point! Here’s another benchmark to help answer that 
-------------------------------------------------
nench.sh v2019.03.01 -- https://git.io/nench.sh
benchmark timestamp: 2019-05-08 10:41:30 UTC
-------------------------------------------------
Processor: QEMU Virtual CPU version 2.5+
CPU cores: 8
Frequency: 2399.996 MHz
RAM: 23G
Swap: -
Kernel: Linux 4.15.0-48-generic x86_64
Disks:
sda 1.5T HDD
CPU: SHA256-hashing 500 MB
3.069 seconds
CPU: bzip2-compressing 500 MB
5.132 seconds
CPU: AES-encrypting 500 MB
1.552 seconds
ioping: seek rate
min/avg/max/mdev = 35.9 us / 62.0 us / 9.43 ms / 75.9 us
ioping: sequential read speed
generated 9.85 k requests in 5.00 s, 2.41 GiB, 1.97 k iops, 492.6 MiB/s
dd: sequential write speed
1st run: 368.12 MiB/s
2nd run: 279.43 MiB/s
3rd run: 354.77 MiB/s
average: 334.10 MiB/s
IPv4 speedtests
your IPv4: 188.68.49.xxxx
Cachefly CDN: 89.13 MiB/s
Leaseweb (NL): 83.95 MiB/s
Softlayer DAL (US): 10.37 MiB/s
Online.net (FR): 13.00 MiB/s
OVH BHS (CA): 3.43 MiB/s
IPv6 speedtests
your IPv6: 2a03:4000:6:xxxx
Leaseweb (NL): 60.51 MiB/s
Softlayer DAL (US): 0.00 MiB/s
Online.net (FR): 28.20 MiB/s
OVH BHS (CA): 10.37 MiB/s
-------------------------------------------------
3 Likes
Very decent !
Will pick up one (or two)
3 Likes
Thank you! Does anyone know whether Netcup allow nested virtualisation on it?
1 Like
I believe @Falzo can confirm, but I remember he saying that the flags are not enabled by default and you need to pay a monthly fee per core.
That’s true, you have to pay.
I had one in the past, what a beast.
yes, that’s my understanding as well, haven’t heard differently.
never used it though at I think it’s 2€ per month and vcore, so for this offer it would roughly triple the costs…
2 Likes
Well, they sell dedicated cores, which do not allow mining and cpu passthrough costs additional 2EUR per core. Which kinda sounds like, they are selling you vcores inofficially.
For what it’s worth, the root servers seem to be cores, rather than vCores. I’d only get their root servers at this point- I had a couple G5/G6 (can’t remember) VPS and ended up letting them just go away because they were ~ 10% steal and the access time was slow. I ran my entire personal web infrastructure briefly on a 3G root server with no issues; for a spell I even dedicated the unused cores to the LET RC5 project.
Here’s the Unix Bench result:
Unix Bench results
Benchmark Run: Wed May 08 2019 15:12:26 - 15:41:00
8 CPUs in system; running 1 parallel copy of testsDhrystone 2 using register variables 35602704.9 lps (10.0 s, 7 samples)
Double-Precision Whetstone 4025.4 MWIPS (12.9 s, 7 samples)
Execl Throughput 2898.2 lps (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 754439.1 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 204632.5 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 2230840.0 KBps (30.0 s, 2 samples)
Pipe Throughput 1089747.6 lps (10.0 s, 7 samples)
Pipe-based Context Switching 59335.9 lps (10.0 s, 7 samples)
Process Creation 6314.2 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 8739.5 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 4249.0 lpm (60.0 s, 2 samples)
System Call Overhead 856455.4 lps (10.0 s, 7 samples)System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 35602704.9 3050.8
Double-Precision Whetstone 55.0 4025.4 731.9
Execl Throughput 43.0 2898.2 674.0
File Copy 1024 bufsize 2000 maxblocks 3960.0 754439.1 1905.1
File Copy 256 bufsize 500 maxblocks 1655.0 204632.5 1236.5
File Copy 4096 bufsize 8000 maxblocks 5800.0 2230840.0 3846.3
Pipe Throughput 12440.0 1089747.6 876.0
Pipe-based Context Switching 4000.0 59335.9 148.3
Process Creation 126.0 6314.2 501.1
Shell Scripts (1 concurrent) 42.4 8739.5 2061.2
Shell Scripts (8 concurrent) 6.0 4249.0 7081.7
System Call Overhead 15000.0 856455.4 571.0
========
System Benchmarks Index Score 1181.5
Benchmark Run: Wed May 08 2019 15:41:00 - 16:09:37
8 CPUs in system; running 8 parallel copies of testsDhrystone 2 using register variables 241108672.5 lps (10.0 s, 7 samples)
Double-Precision Whetstone 33505.5 MWIPS (12.0 s, 7 samples)
Execl Throughput 25879.9 lps (29.8 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 833464.0 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 229442.1 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 2472547.5 KBps (30.0 s, 2 samples)
Pipe Throughput 8177771.4 lps (10.0 s, 7 samples)
Pipe-based Context Switching 887955.8 lps (10.0 s, 7 samples)
Process Creation 49184.8 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 54852.9 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 7213.6 lpm (60.0 s, 2 samples)
System Call Overhead 5606099.9 lps (10.0 s, 7 samples)System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 241108672.5 20660.6
Double-Precision Whetstone 55.0 33505.5 6091.9
Execl Throughput 43.0 25879.9 6018.6
File Copy 1024 bufsize 2000 maxblocks 3960.0 833464.0 2104.7
File Copy 256 bufsize 500 maxblocks 1655.0 229442.1 1386.4
File Copy 4096 bufsize 8000 maxblocks 5800.0 2472547.5 4263.0
Pipe Throughput 12440.0 8177771.4 6573.8
Pipe-based Context Switching 4000.0 887955.8 2219.9
Process Creation 126.0 49184.8 3903.6
Shell Scripts (1 concurrent) 42.4 54852.9 12937.0
Shell Scripts (8 concurrent) 6.0 7213.6 12022.6
System Call Overhead 15000.0 5606099.9 3737.4
========
System Benchmarks Index Score 5103.3
I agree. wouldn’t expect the term ‘dedicated’ means a physical core pinned to a virtual core in 1:1 at all with any provider regardless their marketing blabla